1. Field of the Invention
The present invention relates to color processing of moving image data.
2. Description of the Related Art
Various color perception models have been proposed to estimate human visions of colors. As a typical color perception model, a color appearance model (CAM) called CIECAM02 is known. In the CIECAM02, viewing environment parameters such as a white point, adaptation luminance, background luminance, viewing environment condition, and the like can be set. Upon application of color matching using the CIECAM02, these viewing environment parameters can be set respectively for viewing environments of the input and output sides, and color matching can be done in situations with different viewing environments of the input and output sides.
Also, an image appearance model (iCAM) in which the CAM incorporates the visual spatial characteristic has been proposed. The iCAM executes color matching not only for a still image but also for a moving image, and a visual temporal characteristic is also considered in estimation of appearance of colors of a moving image.
If a point of an image in which the user has an interest (point of interest) can be specified, the CIECAM02 can optimize the adaptation luminance in association with the point of interest. Likewise, in color matching using the iCAM, by optimizing the adaptation luminance in association with the point of interest, the color appearance characteristic can be improved, and color matching closer to human vision can be attained in compression processing of a high-dynamic range (HDR) image.
However, in a moving image, the size of an object corresponding to the point of interest changes due to movement of the object, zooming, and the like. Conventionally, the size of a size-changing object of which the viewer shows interest is not considered. As a result, color matching processing regardless of the point of interest is normally executed.